DESCRIPTION: Work disability is a major burden of rheumatoid arthritis (RA). The most cited study of RA work disability was conducted 15 years ago. Since then, remarkable changes in the macro-economy, in the nature of work, in governmental income support programs, and in the treatment of the disease have occurred. As these are likely to have affected RA work disability, new and more sensitive data are clearly needed. The study's primary aim is to assess work disability among employed persons with RA using a prospective study design. The secondary aims are to assess RA work disability in two increasingly important subgroups, retirement aged persons and minorities, and to assess current risk factors for RA work disability. Risk factors of particular interest are work characteristics that reflect the changing nature of work and consistent use of disease modifying anti-rheumatic drugs. We will study two measures of work disability, work cessation and limitation in ability to work. Subjects will be drawn from patients with RA participating in the National Data Bank for Rheumatic Diseases, a large, national cohort of arthritis patients. 4615 employed subjects with RA aged 18 through 67 will form the full sample; subsets of subjects will be used for some specific aims. Additional patients will be recruited into the NDB cohort to ensure that newly diagnosed patients are included. Subjects will be assessed at the beginning of the project and yearly thereafter for three years for both work disability and factors affecting it. Work cessation is any work stoppage < age 65 and that related to RA > age 65. Limitation in ability to work will be measured using the Work Limitations Questionnaire (WLQ). Incidence of work cessation will be estimated using the Kaplan-Meier method. Trends over time in the WLQ will be examined using random effects regression models. The same methods will be used to examine work disability among retirement-aged subjects. ANCOVA and Cox regression will be used to examine differences in work disability between minorities and Whites. Cox regression, mixed-effects and 1ogistic regression analyses will assess the strength of risk factors. Because of large sample sizes, power estimates indicate results can be determined with fine precision. The information gained will be used for resource allocation, for determining which groups of persons with RA have special needs, and for designing targeted interventions to help persons with RA maintain employment.